Bo An

The growing scale of ad auctions on online advertising platforms has intensified competition, making manual bidding impractical and necessitating auto-bidding to help advertisers achieve their economic goals. Current auto-bidding methods have evolved to use offline reinforcement learning or generative methods to optimize bidding strategies, but they can sometimes behave counterintuitively due to the black-box training manner and limited mode coverage of datasets, leading to challenges in understanding task status and generalization in dynamic ad environments. Large language models (LLMs) offer a promising solution by leveraging prior human knowledge and reasoning abilities to improve auto-bidding performance. However, directly applying LLMs to auto-bidding faces difficulties due to the need for precise actions in competitive auctions and the lack of specialized auto-bidding knowledge, which can lead to hallucinations and suboptimal decisions. To address these challenges, we propose a hierarchical Large auto-Bidding Model (LBM) to leverage the reasoning capabilities of LLMs for developing a superior auto-bidding strategy. This includes a high-level LBM-Think model for reasoning and a low-level LBM-Act model for action generation. Specifically, we propose a dual embedding mechanism to efficiently fuse two modalities, including language and numerical inputs, for language-guided training of the LBM-Act; then, we propose an offline reinforcement fine-tuning technique termed GQPO for mitigating the LLM-Think's hallucinations and enhancing decision-making performance without simulation or real-world rollout like previous multi-turn LLM-based methods. Experiments demonstrate the superiority of a generative backbone based on our LBM, especially in an efficient training manner and generalization ability.

Building intelligent agents capable for autonomously perceiving, reasoning, and acting to achieve goals has been a central pursuit of artificial intelligence (AI) since its inception. For decades, the notion of agency was dominated by symbolic architectures that represent information as formal knowledge and use deliberative reasoning to derive rational actions, offering reliability but at the cost of brittleness and limited generalizability. The recent advancements in large language models (LLMs) and their integration into tool-using, environment-interacting “agentic” systems have reignited interest in AI agents. However, while LLM-based agents provide the flexibility that symbolic systems lacked, they introduce new challenges in reliability and control. We posit that the future of AI agents lies not in indefinitely scaling the model size, but in synthesizing the methods and theories developed by the autonomous agents and multiagent systems community with modern neural architectures to create neuro-symbolic agents capable of trustworthy autonomy.